Analysing the Efficacy of Machine Learning and Deep Learning Models in Breast Cancer Diagnosis
Srinivasan Suganya, A. Ramathilagam · 2025
Breast cancer (BC) affects a large number of female cancer patients worldwide, and it is associated with a high mortality rate. The purpose of this review was to offer various techniques to studying the potential of deep learning (DL) and machine learning (ML) algorithms for early breast cancer detection. Analysing previous studies can assist doctors or medical practitioners in making more accurate diagnoses and providing better care for patients. The article describes many MLAs and DLAs that have been routinely used to detect cancerous cells in BIs early on and classify them based on their malignant status. Also included are current difficulties and potential answers to these issues related to the use of MLAs and DLAs in BC identification. Cancer treatment's primary goal is to eliminate cancer cells. Treatments can reduce tumours to the point where they are undetected by tests. Even after therapy has ended, these injured cells may remain in the body. They once again begin to expand and reproduce over time. Some breast cancers recur even after a lengthy treatment phase has concluded. The investigation discusses the algorithms' techniques, performance, benefits, and drawbacks as they pertain to their use in breast cancer diagnosis. Optimal KNN, hybrid transformer UNet, and transfer learning models fused with LSTM are just a few examples of the ways these technologies have shown promise in improving diagnostic accuracy.